Gastrointestinal Quality of Life Index: Development ...

1 British Journal of Surgery 1995,82,216-222 Gastrointestinal Quality of life index : Development , validation and application of a new instrument E. EYPASCH, J. I. WILLIAMS*, S. WOOD-DAUPHINEEt, B. M. URE, C. SCHMULLING, E. NEUGEBAUER and H. TROIDL II Department of Surgery, University of Cologne, Krankenhaus Koln-Merheim, Koln, Germany, *Clinical Epidemiology Unit, Sunnybrook Health Science Centre, University of Toronto, Ontario, and tSchool of Physical and Occupational Therapy and Department of Epidemiology and Biostatistics, McGill University, Montreal, Quebec, Canada Correspondence to: Dr E. Eypasch, II Department of Suigery, University of Cologne, Krankenhaus Koln-Merheim, Ostmerheimer Strasse 200, 0-51109, Koln, Germany At present, an instrument for measuring the Quality of life , specifically for patients with Gastrointestinal disease, is not available.

2 A new instrument for Gastrointestinal disorders that is system-specific has been developed in three phases. In the first phase, questions were collated and then tested on 70 patients with Gastrointestinal diseases and those that worked well were retained. In the second phase, the questions were modified and tested on 204 patients and the results verified by international experts. The instrument was also validated against other generic measures of Quality of life . During the third phase, the instrument was validated with 168 normal individuals. Reproducibility was tested on 25 patients with stable Gastrointestinal disease and responsiveness was tested on 194 patients undergoing laparoscopic cholecystectomy. The result is a bilingual (German and English) questionnaire containing 36 questions each with five response categories. The responses to questions are summed to give a numerical score.

3 It is concluded that the Gastrointestinal Quality of life index (GIQLI) is ready to be used in clinical practice and research. The question How are you is a central component of the doctor-patient interview. The answer reflects the general well-being or health-related Quality of life of the patient. Today, innovative treatments such as endoscopic surgery or lithotripsy techniques can no longer be evaluated unless facets of the patient s Quality of life are taken into account 1-4. Quality of life is a multidimensional construct with several dimensions: emotional or psychological well being, physical functioning, social functioning, and symptoms of the disease and treatment - . Clinicians, especially surgeons, are sceptical about soft data such as patients reports of symptoms or well-being. Their professional training with an emphasis on pathophysiology leads them to prefer hard clinical data such as laboratory values, imaging results, and survival curves.

4 Techniques are available to harden soft data to make them more manageable and useful*- . A team of four surgeons and three methodologists used these methods to develop and validate a new instrument, the Gastrointestinal Quality of life index (GIQLI). The original idea was to have a set of core questions applicable to any patient with Gastrointestinal disease and several small organ- specific modules. Dependent upon the site of the impairment of the patient, one of the modules is selected for use with the core questions. The GIQLI was developed in three consecutive phases ( Table I). The first phase of Development has been reported in detail elsewhere *. This paper reports the second and third phases of the Development and validation . EDITORS FOOTNOTE: This paper reports a questionnaire previously described in German in a very similar article (Der Chirurg 1993; 64 264-74).

5 The Editors have accepted the argument that parallel Development in English and German represents more than mere translation; the report appears here in order to make the English version widely available Paper accepted 22 June 1994 Table 1 Phases of Development and testing of the Gastrointestinal Quality of life index (GIQLI) Phase I: Question selection, testing and reduction - construction of a questionnaire by the study team - application of the questionnaire to patients, relatives - analysis of the results according to response prevalence - modification and shortening of the questionnaire preliminary testing - application of the modified questionnaire to patients - analysis of the results according to response prevalence - further modification of the questionnaire by the study - definition of scaling and scoring - validation with Spitzer Quality of life index , and health care professionals and impact on Quality of life Phase 11.

6 Continued Development , content verification, and and inter-item correlations team Bradburn Affect Balance Scale and functional activity level gastroenterological experts - evaluation of the questionnaire by 42 Phase 111: Assessment of the measurement properties of the GIQLI - repeated application of the questionnaire in clinically - use with normal individuals to assess validity - assessment of patients with symptomatic gallstones stable patients to test reliability before and after laparoscopic cholecystectomy to test responsiveness cholecystectomy scores used to assess internal consistency - preoperative and postoperative laparoscopic Patients and methods In phase 1, a 76-item questionnaire was applied to 70 patients and 53 close relatives 12. The questionnaire asked about symptoms, physical, emotional and social dysfunction related to gastro- 216 Gastrointestinal Quality OF life index 217 intestinal diseases or their treatments.

7 If a symptom or dysfunction was present, the patient indicated its importance in influencing their Quality of life on a four-point scale. In this phase of Development , only problems that were experienced by at least 25 per cent of the patients and had at least a moderate impact on their qx&y oflife wefe setaked for ftlrther assessment. The characteristics of the individuals participating are presented in Table 2. Phase II: Further Development , content ver$cation and preliminary validation From phase I of the study, 32 items emerged that occurred with sufficient frequency and had a moderate impact on the patients lives. Twelve items were added to focus on the social consequences of being ill and the concerns of the care process. Each of the 44 items was scored on a five-point scale denoting the burden of the particular symptom or dysfunction.

8 The total score ranged from 0 to 176 with higher scores representing better Quality of life . Additional questions were included to obtain demographic, clinical and functional information. Between October 1989 and March 1990, a sample of 204 patients, with benign or malignant disorders of the oesophagus, stomach, gallbladder, pancreas, small intestine, colon, and rectum was asked to complete the questionnaire. Patients with disorders of limited impact on Quality of life , such as appendicitis or haemorrhoids, were excluded. Patients were also excluded if they had abdominal hernias or were unable to read German fluently. Approximately 90 per cent of the patients were inpatients. The rest were outpatients. This phase of the study also incorporated limited validation studies. To test validity, the German version of the Spitzer Quality of life (QL) Inde~ ~, ~ and the Bradburn Affect Balance were added to the questionnaire package.

9 Since the Spitzer index is a global measure of Quality of life and the Bradburn Scale assesses emotional well-being, it was hypothesized that these measures should be positively correlated with the GIQLI scores. Pearsons product moments correlations were calculated. Similarly, based on the known group technique as advocated by B~hrnstedt~ , it was postulated that patients with Gastrointestinal disease who were mobile in the community would have a higher score on the GIQLI than those who were housebound or bedridden. Information on functional activity levels was collected to test this hypothesis. Hospital residents, junior staff members, or trained interviewers explained the questionnaire to the patients, obtained informed Table 2 Characteristics of individuals participating in the three phases of Development and testing of the Gastrointestinal Quality of life index (GIQLI) Mean (range) Per cent Phase Procedure Participants No.

10 Age (years) women -. 70 52(18-86) 44 53 - Professionals 25 - - - I1 Question testing, Patients reduction, validation verification Gastroenterologists Surgeons 1 General practitioners Content I11 Assessing measurement properties Reliability Patients Validity Normal subjects Responsiveness Patients Internal Patients consistency 204 58/22-82) 45 - 24 - 12 - - 6- - 25 50133-78) 45 168 42/25-60) 45 194 50(20-82) 82 194 50120-82) 82 consent, left a copy of the questionnaire, and requested that they complete it. These persons were not directly involved in the care of the particular patient. Questionnaires were collected within 48 h. The final step in phase I1 was to verify the content of the questionnaire with potential users. After evaluating the responses of the 204 aatieffts, a revised versieR v fbe GI & esfttainkg 24 eore questions and eight organ-specific questions was presented to 42 surgeons, physicians and other health professionals from Canada, England, Germany and the USA.